Refolk
November 1, 2026·9 min read

Workday Cut 142 in Pleasanton. Search 'XpressO' in Quotes, Not 'Workday'

Workday's WARN lists 142 Pleasanton tech layoffs effective Nov 30. Find the real engineers with quoted insider terms, not the 71,492 customer profiles.

Workday layoffs 2026Workday layoffs Pleasanton WARNhire laid off Workday engineershow to source engineers from a specific companyBoolean search exclude consultants
Workday Cut 142 in Pleasanton. Search 'XpressO' in Quotes, Not 'Workday'

Workday's September 29 WARN notice gave recruiters something you rarely get: 142 product and technology people at one address in Pleasanton, all with the same end date of November 30, 2026. The catch is that typing "Workday" into a sourcing tool mostly brings back people who use Workday, not the people who build it.

What did Workday's Pleasanton WARN notice actually say?

Workday's California WARN filing lists 142 permanent layoffs at 6110 Stoneridge Mall Road in Pleasanton, effective November 30, 2026. The notice is dated September 29. That gives 62 days of notice, which clears California's 60-day rule.

Workday's SEC 8-K describes "a reduction of approximately 2.5% of Workday's current workforce, primarily within Workday's Product and Technology team, and select leased office space reductions." Workday did not publish a headcount. KRON got to roughly 525 people by applying 2.5% to the 21,070 employees listed in the annual report, and said the Pleasanton cuts "all appear to be" in technology. Treat both the 525 and the division as reporting. Neither is a company disclosure.

The numbers in one place:

  • 142 permanent layoffs in Pleasanton, about 27% of KRON's ~525 estimate
  • $65M to $80M in total charges: $40M to $55M in cash for severance and benefits, about $10M in stock compensation, and about $15M in office lease impairment
  • 296 Pleasanton WARN layoffs across two notices in 2026
  • Fourth round of cuts since 2023, by Boardroom Alpha's count

The earlier 2026 notice matters for sourcing. In February, Workday filed for 154 Pleasanton jobs, effective April 6, and that round mostly hit customer support. September appears to be a different group: the people who build the platform. They describe their work in different words, and those words are how you find them.

Why does searching "Workday" return the wrong people?

Workday is software that other companies' employees use every day, so its name shows up on far more customer profiles than employee profiles. In Refolk's index of professional profiles, only 8 of 25 sampled US profiles with "Workday" in the headline actually work there.

The rest of that sample included founders of companies Workday acquired, board members, and academics. None of them is on the November 30 list.

Query (US, Refolk's index)Total profilesCurrent Workday employees in 25-profile sample
Headline keyword "Workday"37,5348 of 25 (32%)
Skill tag "Workday"71,4920 of 25 (0%)
Headline keyword "XpressO" (unquoted)209,7740 of 25 (matches are "Express" companies)
WARN-listed Pleasanton layoffs142n/a (state WARN data)

The 142 affected people make up 0.38% of the headline pool. Twenty-five profiles is a small sample, so read the percentages as a direction rather than a measurement. The direction is clear, though: a company name that is also a product name is a weak filter.

Workday is a tool people use all day, so its name tracks customers, not staff.

Why are skill tags the worst filter for finding Workday employees?

A skill tag tells you who uses the software, not who signs your paycheck. In Refolk's index, 71,492 US profiles list Workday as a skill, and none of the 25 sampled profiles belonged to a current Workday employee.

The most common titles in that pool were executive assistants, a leadership recruiter, and CTOs - people on the customer side who use Workday in their own jobs. The skill-tag pool is 1.9x the size of the headline pool, and the sample suggests the extra volume is customers.

71,492
US profiles tagging "Workday" as a skill in Refolk's index
None of a 25-profile sample currently worked at Workday.

The rule: filter on employer first. Set current or past employer to Workday as a structured field, then use keywords to narrow inside that set. Never start from a skill.

Most sourcing timelines fall apart at this step. On the skill-tag pool, the 142 people you want are 0.20% of the results, so you'd open about 500 profiles to find one. A recruiter with a req open since September does not have weeks to burn on executive assistants.

Which Workday terms are actually internal?

XpressO and OMS are insider vocabulary. Prism and Extend are not: they mark outside consultants about as often as employees. Rank your terms by how internal they are instead of treating all four the same.

Definitions first:

  • XpressO is Workday's proprietary metadata language. Workday's own technology whitepaper names it and says "Workday's foundation is built on Java." Customers don't write XpressO, so engineers who mention it very likely worked on the platform.
  • OMS (Object Management Server) runs inside Apache Tomcat and provides the XpressO runtime. It's internal, but the acronym is badly overloaded (more on that below).
  • Prism is not insider vocabulary. It shows up on outside consultants' profiles about as often as on employees'.
  • Extend is the same story. Workday sells a training course called "Workday Extend for Developers" to customers and partners, which means outside developers can honestly put it on a profile.
TermWho writes itFalse-positive riskHow to use it
"XpressO"Platform engineersLow if quoted, extreme if notPrimary signal, exact match only
"Object Management Server"Platform engineersLowSpell it out, never "OMS" alone
PrismEmployees, partners, consultantsHighSecondary, only with employer filter
ExtendEmployees, trained partnersHighSecondary, only with employer filter

Acquisitions add one more layer. Workday closed its Paradox and Sana acquisitions in fiscal 2026, and engineers from those teams may still describe their work in their old company's stack terms rather than XpressO. Run a separate pass for them using past employer = Paradox or Sana, current employer = Workday.

Why does unquoted "XpressO" return 209,774 strangers?

Fuzzy keyword matching folds "XpressO" into "Express," so a naive search hands you American Express, Local Express, and EXPRESS. In Refolk's index, unquoted "XpressO" matched 209,774 US profiles, 5.6x the size of the "Workday" headline pool, and none of the sample worked at Workday.

209,774
US profiles matched by unquoted "XpressO" in Refolk's index
Top employers were American Express, Local Express, and EXPRESS.

Fuzzy matching helps when you search "engineer" and want "engineering" too. On a proprietary word one letter away from a common one, the same feature buries you. "OMS" has its own version of this problem, since it also stands for Oracle Management Server, order management systems, and a long list of other things.

Two fixes, and you need both:

  1. Exact-match the phrase. Put "XpressO" in quotes and spell out "Object Management Server."
  2. Pin the employer. An exact phrase with no employer filter still picks up people who write about Workday's architecture from the outside.

What Boolean search excludes consultants and customers?

Put quoted internal terms together with an employer filter, then add NOT clauses for consulting language. Start from the string below, which has not been tested against a live database, and spot-check what comes back before you trust it.

("XpressO" OR "Object Management Server")
AND (current OR past employer: Workday)
NOT (consultant OR "implementation partner" OR HRIS)

How to run it as a workflow:

  1. Set employer as a structured filter, not a keyword. A keyword "Workday" lets every customer back in.
  2. Add the quoted internal terms. Leave Prism and Extend out of the first pass.
  3. Exclude consulting language with the NOT clause.
  4. Add location (Pleasanton, Tri-Valley, Bay Area), since the WARN covers one address.
  5. Spot-check 25 results and count how many actually work at Workday. That's the same test that turned up 8 of 25 on the plain headline search, and your version should beat it clearly.
  6. Run the acquisition pass for Paradox and Sana alumni.

One tradeoff in that NOT clause: excluding "HRIS" can also drop Workday engineers who built the HRIS product and say so. If your spot-check looks thin, remove that term and judge the extra results by hand.

How should you time offers around the November 30 separation date?

Line up offers for a December 1 start, but don't push anyone to sign before they've read their separation agreement. The WARN notice sets only the separation date, so affected staff remain employees through November 30.

Severance terms are not public, and some packages may depend on how the employee conducts themselves until that date. Someone who signs with you in October without checking their agreement could put their package at risk. Raise it before they do.

A workable timeline:

  • October to mid-November: outreach, screens, and loops. People are still employed and can take calls.
  • Before any signature: the candidate checks their separation agreement for conditions tied to conduct or to timing.
  • November 30: separation date in the WARN notice.
  • December 1: earliest clean start date.

Expect competition. Pleasanton Weekly reports that Tri-Valley employers have cut more than 400 jobs since July, including Chevron's 178 in San Ramon, so local supply is rising. Platform engineers who know a Java metadata runtime are a different pool from general Tri-Valley layoffs, though, and Bay Area employers will be competing for them.

To keep up with new filings, watch the primary sources rather than waiting for the news. California EDD's WARN report and trackers like californiawarn.com post notices as they come in, and the @WhatLayoff account on X posted the Pleasanton notice. Boardroom Alpha points out that Workday's executive pay metrics leave out restructuring charges. With four cuts since 2023, keep the search saved.

Future notices may also be easier to source from. From January 1, 2027, California's SB 951 requires covered WARN notices to list the affected occupations, the job functions being automated, and the type of technology, under the header "This notice is for a technology displacement." The bill doesn't define "in substantial part," so expect disputes over when it applies. The September 29 notice came before the law, so it didn't have to say anything about AI.

FAQ

How many people is Workday laying off in 2026?

The September 29 California WARN notice lists 142 permanent layoffs in Pleasanton, effective November 30. The 8-K describes about 2.5% of the workforce, and KRON estimates that at roughly 525 people, though Workday published no headcount. An earlier February notice covered 154 Pleasanton jobs, mostly in customer support, for 296 Pleasanton WARN layoffs in total this year.

When can laid-off Workday engineers start a new job?

The cleanest start date is December 1, the day after the November 30 separation date. Affected staff remain employed until then. Severance terms aren't public and may come with conditions, so line up offers for December 1 and ask candidates to review their separation agreement before signing.

How do you source engineers from a specific company without getting its customers?

Filter on current or past employer as a structured field, never as a keyword or skill tag, then narrow with exact-match terms only insiders use. At Workday that means "XpressO" and "Object Management Server" in quotes. Add NOT clauses for consulting language and spot-check 25 results to confirm the hit rate.

Are Prism and Extend good signals for Workday employees?

Only as secondary signals inside an employer filter. Workday trains customer and partner developers on Extend, and both Extend and Prism appear on plenty of consultant profiles. XpressO and the Object Management Server are internal platform terms and much stronger evidence that someone actually worked at Workday.

The data behind this

Try it on the search you came here for

Stop building boolean strings. Just describe the person.

Type one sentence. I plan the search, read GitHub, public LinkedIn and Crunchbase records, and the open web as it is right now, and hand back a ranked list with the reason next to every name.

  1. 01Describe them

    One plain sentence. Role, city, stack, stage, whatever matters to you.

  2. 02I read the web live

    GitHub, public LinkedIn and Crunchbase records, the open web. Not a database that went stale last quarter.

  3. 03You read the shortlist

    Ranked, with the reasoning under every name. Open a profile, ask a follow-up, narrow it down.

  • No boolean, no filters, no seat to buy. One box.
  • Read at search time, so a profile updated yesterday counts today.
  • Every step visible as it runs, every name with its reason.

500 free credits on sign-up. No card, no demo call. See real searches.

Read next